If it’s early May, then it must be time to talk about what student evaluations of teaching are worth. In a recent essay in Slate, Rebecca Schuman claims that student evaluations are “useless” in their current form, because they encourage students to punish rigorous teachers with low scores and mean comments (and, all too often, sexist or racist ones). The article has gotten a lot of attention from academics I know, who have shared their own stories of uninformed and upsetting comments.
Schuman argues that in light of the unreliability of evaluations, it is unjust to base hiring, firing, and promotion on them. This is especially true for graduate students and part-time faculty members, for whom course evaluations are sometimes the sole documented indicator of job performance.
She is right about the injustice of relying so heavily on a faulty measure of teaching quality, but that in itself does not mean that student evaluations of teaching are useless. It just means we need to use them better. (And, in fact, Schuman does not say we should discard the evaluations altogether, though she does recommend making them more accountable by removing students’ anonymity.)
So how can we make evaluations work better to assess and improve our teaching? Here are four ways to start.
1. Don’t make them the only measure of teaching effectiveness.
Because student evaluations are imperfect, they must be supplemented by other measures that, together, can produce a more reliable picture of a teacher’s effectiveness. Ideally, we could look at what students can actually do at the end of a course (or, in educational lingo, use “valid, direct assessment”). Schuman points to how well her students could communicate in German by the end of her course. But even such an assessment does not tell the whole story, because not all of our courses’ goals are readily measurable after 15 weeks of class. We may want to wait to pass judgment on an instructor until we know how his or her students fare in subsequent courses, or how many get into medical school, or how many become valued resources in their communities. We could even take the philosopher Solon’s advice to the extreme, and call no one a good teacher until all of her students are dead.
In the meantime, though, we do need to make decisions regarding faculty evaluation and development. Classroom visits can help us understand the teacher as a performer and facilitator. Assignments can tell us about rigor and creativity. Responses to students’ papers, exams, and lab reports can indicate the teacher’s empathy and ability to pinpoint how students can improve. Small Group Instructional Diagnosis can elicit constructive student feedback while filtering out the extreme student voices.
2. Know the capabilities and limitations of the specific evaluation instrument.
Where possible, employ forms that have been shown to be statistically valid measures of several independent aspects of the instructor and course. The Student Evaluation of Educational Quality (SEEQ) is one example. At King’s College, we moved to the SEEQ after an analysis of our previous instrument showed that it was measuring only one variable: whether the student liked the class or not. In addition to measuring students’ perception of the instructor’s enthusiasm, preparation, and other aspects of his or her teaching, SEEQ also asks students about the relative pace and workload of the course, revealing whether they find the course “hard,” and so giving more context to interpret the raw evaluation scores.
3. Acknowledge that evaluation scores are correlated with students’ expected grades.
Or, perhaps more accurately, acknowledge that they are correlated with expected grades relative to students’ GPAs. Once you do that, you can measure the relationship between scores and grades and adjust the scores accordingly. Teachers whose scores are consistently above the adjusted mean are doing something that students appreciate, and teachers whose scores are consistently below the adjusted mean are doing something that students don’t appreciate. Find out what those things are by talking with them about their teaching, visiting their classrooms, and reading their assignments and comments on student work.
And just to head off one avenue of critique, I’ll say that doing things that students appreciate is not the same as pandering to them. If students feel valued, if they feel comfortable in a class, if they feel supported, if they just like being in the class, then several obstacles to their learning are removed, and they have a better chance of success.
4. Don’t leave student evaluation of teaching for the end of the course.
Student evaluations of teaching are useless if the teacher finds out what students think about the course only after the course is over. If you want to know what students think about the course while there is still time to make adjustments, you can ask them to fill out the evaluation forms at midsemester. (This should be done anonymously, because while grades are still unsettled, students’ worries about retribution have more legitimacy.) The midterm can also be an opportunity to get students to evaluate their learning, so that they can make changes in the second half of the course. Evidence indicates that instructors who solicit and respond to such feedback end up with higher end-of-term evaluations.
Student evaluations of teaching do not tell us everything we’d like to know about ourselves as teachers. And they do permit students to turn nasty. But they do tell us something. By using them as one measure of teaching effectiveness, we do not capitulate to the student-as-consumer model of education. Rather, we put a measure of faith in our students’ sense (if imperfect) of what is good for them. We acknowledge that our courses are about their learning and that we care about how better to enable that.
Jonathan Malesic is an associate professor of theology and director of the Center for Excellence in Learning and Teaching at King’s College, in Wilkes-Barre, Pa.